Nick Cox <[email protected]> and Alan Riley <[email protected]> have both
suggested that Mark Schaffer <[email protected]> follow the lead of the
clickable commands in graphics help files and make each clickable completely
self contained; that is to say wrap the desired command in a block with
preserves, data loading, and any other commands that must be run.
Alternately, each clickable could do exactly what it claims and run the exact
wording of the clickable as a Stata command. The choice was easy in the
graphics help files, we just wanted to let the user draw some graphs. The
graph command itself was what was important, and loading the data was a
side-issue.
Mark has a somewhat different problem. He is demonstrating the process of
doing an analysis, a process that necessarily involves sequences of commands.
It might even be considered deceptive if the clickable for a post-estimation
command automatically loaded the data, generated some variables, ran an
estimation command, and then finally did ran the post-estimation command that
was clicked. Part of the learning process is knowing that some commands must
be run first. Not that it won't be frustrating when one of the clickables
fails if a necessary prior step was skipped or an additional destructive step
was inserted.
Personally, I think there are times when I will want to use both the
"all-in-one" graphics style and the "you-must-follow-the-path" tutorial style.
-- Vince
[email protected]
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
